Thought Experiments = Bad Philosophy
https://aeon.co/essays/what-is-the-problem-with-ethical-trolley-problems
So too, I think, should philosophers. It's hard to imagine more terrible ways of thinking about philosophical problems than via thought experiment. At best, they ought to be used as examples of how not to think; or how to think in circumstances that are extremely constrained and rare.
A recent instance in which the 'trolly problem' become prominent in public discourse I think illustrates this well. In Italy, during the height of their COVID problems, it was said that doctors were being forced to choose between patients - who ought to get treatment, and who should die. Predictably, this had a few commentators likening the issue to the trolley problem. In some ways, the likening was exactly right. However, the situation in Italy shed as much light on the trolly problem as did the other way around: that trolly problems only become problems in instances of acute disaster, when the normal run of things go totally out of whack. To think, on the other hand, that answers to the trolly problem - or answers about who should die - are generalizable into broader ethical principles, is, I think absurd.
That's just one example. The article linked gives some nice intrinsic reasons why thought expriements make for pretty terrible philosophy. Among them of course being that thought experiments are almost uniformly artificial and, again, totally ungeneralizable. The article itself focuses on what it calls ethical thought experiments, but I think the same is true for other well known ones too. The damage that 'brain in the vat' thought experiments have wrought on philosophy of mind, for instance, is I think incalculable. But that's another story.
In any case, this is mostly an excuse to pimp out the article, and induce some discussion about the role of thought experiments in philosophy more generally.
While thought experiments are as old as philosophy itself, the weight placed on them in recent philosophy is distinctive. Even when scenarios are highly unrealistic, judgments about them are thought to have wide-ranging implications for what should be done in the real world. The assumption is that, if you can show that a point of ethical principle holds in one artfully designed case, however bizarre, then this tells us something significant. Many non-philosophers baulk at this suggestion.
So too, I think, should philosophers. It's hard to imagine more terrible ways of thinking about philosophical problems than via thought experiment. At best, they ought to be used as examples of how not to think; or how to think in circumstances that are extremely constrained and rare.
A recent instance in which the 'trolly problem' become prominent in public discourse I think illustrates this well. In Italy, during the height of their COVID problems, it was said that doctors were being forced to choose between patients - who ought to get treatment, and who should die. Predictably, this had a few commentators likening the issue to the trolley problem. In some ways, the likening was exactly right. However, the situation in Italy shed as much light on the trolly problem as did the other way around: that trolly problems only become problems in instances of acute disaster, when the normal run of things go totally out of whack. To think, on the other hand, that answers to the trolly problem - or answers about who should die - are generalizable into broader ethical principles, is, I think absurd.
That's just one example. The article linked gives some nice intrinsic reasons why thought expriements make for pretty terrible philosophy. Among them of course being that thought experiments are almost uniformly artificial and, again, totally ungeneralizable. The article itself focuses on what it calls ethical thought experiments, but I think the same is true for other well known ones too. The damage that 'brain in the vat' thought experiments have wrought on philosophy of mind, for instance, is I think incalculable. But that's another story.
In any case, this is mostly an excuse to pimp out the article, and induce some discussion about the role of thought experiments in philosophy more generally.
Comments (41)
Here is the tram problem in its infancy: http://www2.econ.iastate.edu/classes/econ362/hallam/Readings/FootDoubleEffect.pdf
Interesting that Foot is not so interested in the solution, as to the comparison with the judge and the rioters. It seems to me that the tram problem fuss is a perversion of the original use to which it was put.
And it was a tram, not a trolley. The 'mercan hegemony, again.
Well, I took utilitarianism for prelims at Oxford in 1979, which was way before the trolley problem, and the lecturer to the prelims class posed much the same question, instead of about whether we'd join a posse to shoot a criminal, as part of his debate on act versus rule utilitarianism. I dont know if the trolley problem really helps any better to understand the conflict, because it doesn't actually require understanding the difference between act and rule utilitarianism, and people are led to believe their intuition is more important than learning the far more extensive thinking in prior history than intuition could really achieve by itself for most people Ive met.
I'm not terribly troubled by thought experiments. If one offers an opinion based upon a principle, it's appropriate to hypothesize the application of that principle in novel situations. If it's apparent that the principle fails to provide an adequate solution in those hypothesized situations, you're faced with the choice of reconsidering the principles upon which you rely or in accepting that the principle works only to a point and then delineating those limits.
The observation that many thought experiments are invalid because of hidden distinctions contained within them is true, but that just means the proper way to combat argument by analogy is by pointing out the distinctions that make the analogy disanalagous. The debate will then center on whether those distinctions are relevant to the broader point.
What we're discussing here has real world application and is not limited to the philosopher's armchair. Anglo oriented legal systems rely heavily upon analogy of precedent, deriving principles from prior judicial opinions and applying them to novel situations. In the legal context, the prior and the novel situations are all actual (as opposed to hypothetical), and it's also logically appropriate to ask how the principle enunciated might apply in hypothesized situations (reductio ad adsurdem considerations worked out as well).
The clever lawyer does exactly as the author indicates the informed medical provider does in the article you cited: he explains why the analogy doesn't apply, he points out distinctions, and he explains why the deciphered principle has narrow implications to only very particular situations.
The fact that argument from analogy results in eternal bickering over which facts matter and why yesterday's decision shouldn't apply to today's facts is the nature of the beast, but that doesn't make it useless, especially in the legal context, where we defer to the court as the arbiter of who has got the analogy correct and how it shall be applied. Of course, even then, we fully expect those rulings to be rethought and reconsidered by later courts. I see this way of reasoning productive though, taking full advantage of human's capacity to think deeply into issues and in giving deference to those opinions most well thought out, especially those that have satisfied the most pressing hypothetical thought experiments.
I’ve been over this before quite thoroughly and it is surprising how many people just dig their heels in at any suggestion that the ‘purpose’ they see might just be completely wrong.
Any public proclamation is always biased by the perceived biases in others the proclaimer bring into the public sphere. For me the point of such ethical dilemmas framed in thought experiments (hyperbolic or otherwise) is to first and foremost be honest with oneself rather than curb personal thoughts simply because they’re uncomfortable.
People in the emergency services and the army train in this manner. The same comes into play for us as individuals. It takes work to fortify our ethical positions with actual actions and behaviors that adhere to them. Probably all of us say one thing and do another a lot of the time, but preparation of thought can have us acting more like we’d have wished to rather than simply ignoring the inconvenient truth of our susceptibility to failing to act as we, at our core, truly deem fit.
Quoting StreetlightX
This is missing the point. The more extreme the scenario is serves only to bring up your personal take on the matter. They are opportunities to see why you think what you think, what you’d prefer to think, what you’d say as opposed to what you really think, and what can be done to balance these things ... to name a few paths of enquiry.
Quoting StreetlightX
I’ll have to have a closer look at it. Clearly the use I find they don’t from what you’ve espoused.
I completely understand that a reasonable number of people find them actively repulsive - I just think they’re looking at them in too rigid a fashion.
It's definitely an interesting use of a thought-experiment. I withdraw some of my animus towards the darn thing, but will still note how circumscribed it's use was in the paper - as part of an effort to draw a new distinction.
I have much sympathy for your animosity towards misused trams.
What is it like to be a bat? Nagel.
What is it like to be a a light ray? Einstein.
Is it something about ethics that makes them objectionable?
I would say that ethics suffers from the usual problem of social sciences, that the object of the experiment is altered, not only by the experiment but by the theory which is applied.
For example, protecting one's family is usually considered a strong justification for almost any action. However, in a time of plague, there is a more over-riding priority to stop the spread of disease, even at the cost of one's own or one's family's life. Obligations change, societies change, people change, and the morality of this or that action changes. There is no stable condition equivalent to the speed of light, or the use of echo-location, that can transfer to and from the thought experiment, and allow for universalising.
Also, the p-zombie thought experiment is good for pointing out the difficulty with incorporating consciousness into a material framework. But also the difficulty when you don't, since p-zombie Chalmers is making the exact same argument!
For some reason or none, it never meant anything much to me. Kind of like the trolley, you can make it go which way you want.
A trolley operated by a p zombie is like a self-driving car, passengers or pedestrians? But the zombie has no morality by definition, we have to program it with our morals.
Then there's the bizarre dissonance between on the one hand clearly investigating something we don't currently know whilst on the other presuming we have such flawless knowledge of all the other variables that we can perform this mental trick of holding them steady whilst we conduct our 'controlled trial'.
The p-zombie argument has little to do with ethics, but one might argue that torturing a p-zombie wouldn't be wrong since it doesn't feel pain.
However, what if it turns out we're all p-zombies? Does that mean we get to torture one another? Probably we would adjust our ethics instead. Although I don't know what exactly it would men to say that I don't really experience suffering.
Child: Mummy I had a dream where I was just a big ball of pink goo in a jar!
Parent: That's ok dear, it wasn't real.
Child: It was scary.
Parent hugs child.
Philosopher: I have a thought experiment where I'm just a brain in a jar.
That's ok dear, that's ok.
If you're going to straw man it, sure. But it's just expressing a modern version of age-old concerns about skepticism, because our heads are the jars. How do we know our senses are telling us the truth about the world?
They don't even deserve the name of experiments. It's just certain philosophies attempt to buy into the prestige of scientific association. Thought fantasies is a more appropriate appellation.
Child: Mummy I had a dream where I was just a big ball of pink goo in a jar!
Parent: That's ok dear, it wasn't real.
Child: That's a petitio principi.
Yes, but what if the p-zombie was just a brain in a vat? What if he was colourblind tortured p-zombie brain in a vat? Would he learn anything new when he sees red?
Like Maxwell's demon or Schrödinger's cat?
Like naive realism.
He'd learn he was a bat dreaming of being John Wick in the Matrix.
...
Quoting fdrake
Still an appropriate response, no?
Quoting Marchesk
Yes, but what is it like to be a bat dreaming of being John Wick in the Matrix?
We did all basically agree to Kantianism in that direct realism thread, did we not?
Quoting Isaac
Fucking bad ass, but I'm not sure the tortured, colorblind bat BIV would be able to see this:
And tasty wheat would still probably taste like chicken.
Wouldn't physicalism have the same criticism? Those thought experiments are pointing out the difficulty with explaining everything in physical terms.
Unfortunately for you, some philosophers do.
You stated that it would be better if "we" moved on from these dumb thought fantasies.
The way I see it, that kind of thought experiment is more a tool to check how the moving parts of a philosophy work in extremis. If the result is absurd, that's good cause to check where that absurdity comes from.
Exactly, though that particular example is also a cautionary tale about how easy it is to misunderstand a philosophy if you only look at a thought experiment.
:up: good post.
Can you think of an example of this, where an absurd result from a thought experiment has been a red flag in this sense. I'm not entirely sure what you mean and I think an example might help.
That sums it up for me. Sum people approach them with distain believing they’re meant to guide everyone to sum kind of ‘ethical’ consensus. They are, in terms of ethics, extremely useful for seeking/seeing the nuances of how ‘cold’ reasoning plays its part in shifting the burden of responsible action/thought.
Ethics isn’t merely about exchange figures and summing up some total solution. Thought experiments and hypothetical scenarios are not calculations.
The "lying is always wrong" example Marchesk brought up is commonly used in this way. Let's assume you have a deontological moral philosophy that argues that certain acts are immoral regardless of their consequences, and an example is lying. Then someone brings up the Nazis searching for a Jew hiding in your house example. Many people will find the conclusion that you have to tell the truth to the Nazis, because lying is wrong, absurd.
Faced with this result, you'd have to either figure out why the initial reaction of "that's absurd" is false, or revise your answer.
It's basically like applying a simulated real-world test to check and see if something holds, and they're especially useful if we can rapidly iterate through thought experiments in order to refine a generalization/model/rule/prediction.
Thought experiments are especially bad when the underlying simulation being run (the worldview and supporting premises of the thinker, more or less) is itself bad. Appealing to incorrect ontic/epistemic/physical/metaphysical priors just imbues the thought experiment with the same inaccuracy.
Ultimately, thought experiments (or at least the "rules" they help us refine) must be put to the test in the real world. If they're never actually put to the test, then the thought experiments in question may in fact have been bad and useless all along...
Can we anticipate before hand whether a given area or field of inquiry, and the accompanying thought experiments that would help expedite that inquiry, will turn out to be useless? Coincidentally, my instinct is to use thought experiments to begin exploring that question.
Using very broad and vague terms that are borrowed from reflections on complex systems, thought experiments actually represent a very interesting and important layer of "intelligence", where intelligence is defined as a planning-capable sensory-experience-having agent.
The evolutionary purpose of our very capacity for thought experiment is that it allows us to make projections about cause and effect (it helps us understand the past, make sense of the present, and anticipate the future): the minds eye.
Taken at face value, thought experiments are like imaginary canvases upon which we can noodle and practice, and create anticipatory models. For instance, without this capacity, we could not anticipate the motives and intentions of other intelligent agents, and this is capacity almost directly impacts our capacity to survive and reproduce in social environments.
The more that I think about it, the more the titular conjecture seems patently false. There are lots of bad philosophies that rely heavily on shitty thought experiments; thought experiments alone does not good philosophy make. Valid priors and real world utility really goes a long way. But without thought experiment, without imagination, we wouldn't have the cognitive power to explore more complex ideas in the first place.
For the first time, I whole-heartedly agree. Not that it means anything, but :clap: